We propose an international collaborative multidisciplinary research project to investigate neural circuit mechanisms of spatial memory formation in the mammalian brain. In both humans and rodents, the hippocampus is a brain region required to store long-term memories. This process depends on activity- dependent changes in the strengths of connections between neurons, known as synaptic plasticity. In the input layer of the hippocampus, the dentate gyrus, a principal cell type called the granule cell receives synaptic inputs from cortex carrying information about spatial navigation, and it is thought that synaptic plasticity at these inputs stores a map of space in dentate granule cells. Like most hippocampal and cortical circuits, the dentate gyrus is comprised of multiple neuronal cell types, many of which are inhibitory interneurons. However, the dentate gyrus also contains a unique class of local excitatory interneurons,, called mossy cells, which form local synapses onto granule cells and inhibitory interneurons, but do not project outside the dentate gyrus. It is a major goal of this project to reconcile the direct excitatory and indirect inhibitory influences of mossy cells on granule cells, which could have opposing effects on spatial information processing and memory storage. Recent work has shown that mossy cells are required for the formation of new spatial memories, though the underlying mechanism is completely unknown. Interestingly, the mossy cell inputs and cortical inputs to granule cells segregate onto proximal and distal regions of granule cell dendrites, respectively. In other cell types, synchronous activation of proximal and distal input pathways evokes local dendritic spikes that potently induce synaptic plasticity. However, whether this occurs in granule cells in response to coincidence of feedforward input from cortex and feedback input from mossy cells has not been investigated. Therefore, in order to test the role of mossy cell input in gating dendritic spiking and plasticity in dentate granule cells, we propose to 1) directly record from granule cell dendrites in vitro in response to precisely controlled input patterns, 2) directly image the activity of granule cell dendrites in vivo during spatial navigation while chemogenetically silencing mossy cells, and 3) develop experimentally-constrained computational models of dentate gyrus cells and circuits to investigate how a dedicated feedback neuron type like mossy cells affects the storage and recall of information in a neural circuit. RELEVANCE (See instructions): Humans with temporal lobe epilepsy and traumatic brain injury exhibit specific degeneration of hippocampal mossy cells, and suffer from deficits in memory and cognition. While existing pharmacological treatments reduce seizure frequency and severity in some epileptic patients, no treatments exist for the associated learning and memory disorders. We expect our research to generate new insight and inform the development of new treatment strategies for neural circuit dysfunction.